Double Bundle Method for Nonsmooth DC Optimization
نویسندگان
چکیده
The aim of this paper is to introduce a new proximal double bundle method for unconstrained nonsmooth DC optimization, where the objective function is presented as a difference of two convex (DC) functions. The novelty in our method is a new stopping procedure guaranteeing Clarke stationarity for solutions by utilizing only DC components of the objective function. This optimality condition is stronger than the criticality condition typically used in DC programming. Moreover, if a candidate solution is not Clarke stationary, then the stopping procedure yields a descent direction. With this new stopping procedure we can avoid some drawbacks, which are encountered when criticality is used. The finite convergence of the method is proved to a Clarke stationary point under mild assumptions. Finally, some encouraging numerical results are presented.
منابع مشابه
Limited memory interior point bundle method for large inequality constrained nonsmooth minimization
Many practical optimization problems involve nonsmooth (that is, not necessarily differentiable) functions of hundreds or thousands of variables with various constraints. In this paper, we describe a new efficient adaptive limited memory interior point bundle method for large, possible nonconvex, nonsmooth inequality constrained optimization. The method is a hybrid of the nonsmooth variable met...
متن کاملA bundle filter method for nonsmooth nonlinear optimization
We consider minimizing a nonsmooth objective subject to nonsmooth constraints. The nonsmooth functions are approximated by a bundle of subgradients. The novel idea of a filter is used to promote global convergence.
متن کاملLMBM — FORTRAN Subroutines for Large-Scale Nonsmooth Minimization: User’s Manual
LMBM is a limited memory bundle method for large-scale nonsmooth, possibly nonconvex, optimization. It is intended for problems that are difficult or even impossible to solve with classical gradient-based optimization methods due to nonsmoothness and for problems that can not be solved efficiently with standard nonsmooth optimization methods (like proximal bundle and bundle trust methods) due t...
متن کاملLimited memory bundle method for large bound constrained nonsmooth optimization: convergence analysis
Practical optimization problems often involve nonsmooth functions of hundreds or thousands of variables. As a rule, the variables in such large problems are restricted to certain meaningful intervals. In the paper [Karmitsa, Mäkelä, 2009] we described an efficient limited memory bundle method for large-scale nonsmooth, possibly nonconvex, bound constrained optimization. Although this method wor...
متن کاملLimited Memory Bundle Method for Large Bound Constrained Nonsmooth Optimization
1. Abstract Practical optimization problems often involve nonsmooth functions of hundreds or thousands of variables. As a rule, the variables in such large problems are restricted to certain meaningful intervals. In the report [Haarala, Mäkelä, 2006] we have described an efficient adaptive limited memory bundle method for large-scale nonsmooth, possibly nonconvex, bound constrained optimization...
متن کامل